Reading The Medical Literature Lecture Slides PDF

HREP5104 READING THE MEDICAL LITERATURE THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ Learning Objectives On completion of the required readings and this session, you should be able to Review common study designs and their alignment with the research question Evaluate the met...

HREP5104 READING THE MEDICAL LITERATURE THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ Learning Objectives On completion of the required readings and this session, you should be able to Review common study designs and their alignment with the research question Evaluate the methodological quality of a scientific publication and use the appropriate methodological template Critique the scientific literature addressing a specific research question Assess the quality of the presentation of the information Identify potential sources of bias and confounding in a scientific publication Reconcile the strengths and weaknesses of a scientific publication THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Introduction OUTLINE Review of Study Designs Critique of a Scientific Paper THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Why Should We Read Scientific Literature? THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Critical Appraisal of Research Article “a systematic process used to identify the strengths and weaknesses of a research article in order to assess the usefulness and validity of the research findings”. Important components include: Appropriateness of the study design for the research question Key methodological features Statistical analysis How to critically appraise an article Jane M Young 1, Michael J Solomon THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  EPIDEMIOLOGICAL STUDIES Observational (non- Interventional experimental) (Experimental) Data Data from Data Data from from individual from individual groups s groups s Descripti Analytic Descripti Analytic ve ve Clinical trial Ecologica Cross- Cohort Case-control Communit Individual l study sectional study study y trial field trial study Figure THE UNIVERSITY OF 1:INDIES THE WEST Quantitative | CAIHR Epidemiological www.uwi.edu/  Observational Studies THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Cross-Sectional Study Examines the relationships between health outcomes and other characteristics of a defined population at a particular point in time Provides the prevalence of disease, exposure and/or risk factors Good for improving disease awareness Public Health Assessment and planning of health intervention and policy THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Cross-Sectional Study Advantages Quick Cannot Disadvantages determine Inexpensive causation May be useful to plan other studies THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Case Control Study Patients with a particular disease or condition (cases) are identified and their past exposure compared with persons who don’t have the disease (controls) Investigate hypotheses on the etiology of disease ( association of risk factors with the disease) Case – clearly defined; may be based on clinical findings, laboratory or other investigations or a combination Controls – should not have outcome of interest or high/low prevalence of the exposure THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Case Control Study Advantages Disadvantages Multiple exposures can be studied Temporality cannot be simultaneously determined Study rare diseases Not useful for rare exposures Study diseases with long latency period Cannot study multiple outcomes Cost effective Cannot directly compute Can be conducted quickly incidence or prevalence THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Cohort Study The cohort is any group of persons who share a common characteristic Persons are classified (or sometimes selected) according to exposure to a particular factor and followed until the development of a disease or outcome Cohorts can be retrospective or prospective Retrospective cohort is also called a historical prospective study, nonconcurrent prospective study or a prospective study in retrospect Prospective cohort is also called a longitudinal study, follow up, incidence or a concurrent prospective study Remember in cohort studies subjects are always selected on the basis of their exposure THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Cohort Study Advantages Disadvantages Can establish a temporal relationship between Expensive – large numbers and long follow exposure and outcome up time usually required Can assess the effects of rare exposures on disease risk Time consuming – will investigator survive Can examine multiple effects from a single to see results? exposure Can assess the effect of multiple exposures on the Prone to bias from loss to follow up, non- same outcome participation Less prone to selection bias (though this may be Not useful for rare diseases an issues in retrospective cohorts), information bias Can be limited by available records THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  EXPERIMENT AL STUDIES THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Clinical Trials In this type of study, the investigator administers the exposure to the subject (clinical trial) or to a group community (a community intervention or cluster trial) Most expensive study design but often the most definitive type of study in establishing a cause-and-effect relationship There are several ethical issues involved with the administration of an exposure The most important one is that clinical equipoise exist i.e. there must be some degree of uncertainty about the benefits or harms that may result from this intervention THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Randomized Controlled Trials Randomization - allows for an equal distribution of known and unknown risk factors between the intervention and control / placebo group. Randomization, if properly conducted, removes confounding Each subject is carefully selected and screened The intervention is randomly assigned All subjects are carefully followed throughout the study to the time of the outcome of interest or the end of the study THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Clinical Trials Strengths Limitations Proves the effect or lack of Can be affected by failure of effect from a given randomization or poor methods for randomization exposure More expensive and time Provides the highest level of consuming evidence of a causal Problems can occur if protocol not association carefully followed THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  CLASS ACTIVITY State the study design that was used in the following articles Sleep Problems in Children with Autism Spectr um Disorder: A Multicenter Survey Risk Factors for Acute Kidney Injury During A minoglycoside Therapy in Patients with Cystic Fibrosis Language Interpretation Conditions and Boun daries in Multilingual and Multicultural Emerg ency Healthcare Once-Weekly Semaglutide in Adolescents with Obesity THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  BREATHE STRETCH TAKE A 1 MIN BREAK THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Bias & Confounding THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Possible Explanations for Association Association Found True/Causal Random Association Bias Confounding Error 25 THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Bias Definition “any systematic error in an epidemiological study that results in an incorrect estimate of the association between exposure and risk of disease” Bias results from systematic errors in the research methodology Two main types: Selection Information THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Bias – The Nemesis of Validity Primarily affects internal validity ‘All observational studies… have built- in bias; the challenge for investigators, editors, and readers is to ferret these out and judge how they might have affected results.’ Source: Grimes D & Schultz K. Bias and causal associations in observational research. Lancet. 2002;359:248-52 27 THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ Potential Sources of Bias in a Cross- Sectional Study Selection Bias – usually affected by sampling strategy Observer Bias – errors in measurement Non-response Bias – non-representative sample Recall Bias – retrospective, external or internal factors THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ Potential Sources of Bias in a Case Control Study Recall Bias - cases and controls recall past exposure differently Selection Bias - when those individuals selected as controls are unrepresentative of the population that produced the cases. Interviewer Bias -recording exposure information may be influenced by the investigator's knowledge of an individual's disease status Ascertainment Bias - when there is inaccurate ascertainment of either the disease or exposure of interest. THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Potential Sources of Bias in a Cohort Study Diagnostic bias – assessment of outcome Information Bias – misclassification of subjects according to exposure and outcome Non-response Bias Selection Bias THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Potential Sources of Bias in a Cohort Study Misclassification – the erroneous classification of an individual, a value or attribute into a category to which they should not have been assigned. Two (2) types of misclassification differential - does not occur at random non-differential - occurs at random THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Potential Sources of Bias in a Clinical Trial Selection Bias – incomplete randomization Performance Bias – differences in level of care provided Exclusion Bias – differences in withdrawals or loss to follow up Detection Bias – differences in outcome assessment Publication Bias – selective reporting of findings THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ How Do We Deal with the Main Types of Error in Research Studies? Random Error Confounding Bias (Chance) Perform hypothesis test to estimate Control for Avoid – through the role of chance confounding using good design and special methods conduct of study Calculate (randomization, confidence standardization, intervals to give an stratified analysis) estimate of the range of possible values THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Confounding A confounder is an exposure, external to our hypothesis, that biases our measure of association unless it is controlled. Usually, confounding will attenuate or exaggerate the association (sometimes called negative and positive confounding). Last, 1995 THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Properties of a Confounder Exposure Outcome Confounder A risk factor for the disease i.e. must be associated with outcome of interest Associated with the exposure and the outcome independently Not be an intermediate step between exposure and outcome (not on causal pathway) 35 THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Confounding – Examples Alcohol Lung Cancer Consumption Smoking status 36 THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  METHODOLOGICA L TEMPLATES THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ This Photo by Unknown Author is licensed under CC BY-  STROBE Checklist The STROBE Statement is a checklist of items that should be included in reports of observational studies. THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ STROBE Checklist THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  CONSORT Checklist The CONSORT Statement is a checklist of items that should be included in reports of randomised trials. THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  CONSORT Checklist THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ What is the Best Study Design for Each Specific Research Question? What is the prevalence of asthma in Jamaican children aged 2-17 years? The incidence of thyroid cancer in Chinese patients after childhood irradiation for keloid scars Myocardial infarction and the use of estrogen and estrogen- progesterone in post-menopausal women Effect of hydroxychloroquine on COVID-19 prevention in cancer patients undergoing treatment THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  The Best Study Designs are… What is the prevalence of asthma in Jamaican children aged 2-17 years? Ans: Cross-sectional study The incidence of thyroid cancer in Chinese patients after childhood irradiation for keloid scars Ans: Cohort study Myocardial infarction and the use of estrogen and estrogen-progesterone in post-menopausal women. Ans: Case control study Effect of hydroxychloroquine on COVID-19 prevention in cancer patients undergoing treatment Ans: Clinical trial THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Critique of the Literature 1.Statement of Problem and Purpose 2.Review of the Literature 3.Search Strategy * 4.Study Design * 5.Research Question and Relevance 6.Description of the Methods 7.Findings of Critical Appraisal Questions of Validity 8.Summary of the Primary Results 9.Assessment of External Validity 10.Conclusion THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/ Statement of Problem and Purpose - (Background) RATIONALE IMPORTANCE OF ORIGINALITY OF THE PROBLEM THE STUDY THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Review of the Literature Includes: Search Strategy Study Design Research question and relevance PICO/PEO/PICOT THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Description of the Methods Selection of participants Measurements of outcome and exposure Statistical analysis Methods to prevent bias Example: “This paper describes an unblinded randomised trial, concerned with therapy, in 267 hospital outpatients aged between 58 and 93 years, in which four-layer compression bandaging was compared with standard single-layer dressings in the management of uncomplicated venous leg ulcers. Follow-up was six months. Percentage healing of the ulcer was measured from baseline in terms of the surface area of a tracing of the wound taken by the district nurse and calculated by a computer scanning device. Results were analysed using the Wilcoxon matched pairs test.” Trisha Greenhalgh, How to Read a Paper: The Basics of Evidence-based Medicine and Healthcare, 5th Edition THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Findings of Critical Appraisal Questions of Validity Utilize appropriate tool for evaluation of study design STROBE CONSORT The role of potential bias or confounding THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Summary of the Primary Results 1 2 3 Summarize Report secondary Mention any results in relation results (if significant to the research significant or findings in the question relevant) figures or tables. primarily THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Assessment of External Validity Have the authors provided defensible evidence of the generalizability of the findings? Determine the applicability to a particular target population and your practice or setting. THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  Conclusion State conclusions about the findings State the resolution to the research question State implications of the study Provide recommendations THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/  References Greenhalgh, T How to Read a Paper: The Basics of Evidence-Based Medicine, 5th Edition. BMJ Books. Wiley-Blackwell 2012 Young, J. M., & Solomon, M. J. (2009). How to critically appraise an article. Nature Clinical Practice Gastroenterology and Hepatology, 6(2), 82. THE UNIVERSITY OF THE WEST INDIES | CAIHR www.uwi.edu/